Intermolecular interactions that define the specificity and timing of cellular signaling pathways are driven by a relatively limited set of protein and lipid interaction domains, which have been the focus of previous project periods. In the last funding cycle. we made key discoveries that motivate the two Specific Aims that constitute the revised research plan. Both are poised for important short-term progress with direct relevance for understanding and attacking human disease, and require new personnel for their progress and completion. In the first, a search for novel lipid-binding domains in yeast has led us to identify the 'kinase associated-1', or KA-1 domain as a membrane targeting domain that is important for regulating a series of important kinases in yeast and - importantly - in humans. Prior to our recent discovery, the KA-1 domain was the only example among the common Pfam domains in human kinases for which the function was not known. It is found at the C-terminus of the MARK (MAP/microtl,Jbule affinity-regulating kinase)/Par-1 family of kinases in humans, which are involved in cell cycle. cell polarity and signaling control. MARK kinases phosphorylate microtubuleassociated proteins (MAPs) and the Tau protein, and have been shown to play an important role in axondendrite specification. Alterations in expression or activity have been implicated in autism and Alzheimer's disease - where Tau is hyperphosphorylated by MARK kinases. Our data identify the C-terminal membranetargeting KA-1 domain as an important determinant of subcellular localization and activity. Understanding the mechanism of this regulation in Aim 1 will open new avenues for therapeutic manipulation of MARK kinases in Alzheimer's disease. Our second important discovery was that many pleckstrin homology (PH) domains, generally assumed all to bind phosphoinositide lipids, recognize target sequences in proteins that 'mimic'phosphoinositides by presenting 2 or 3 acidic groups in a particular arrangement. PH domains are the 11 th most frequent protein domains in the human proteome, play crucial roles in a wide range of interactions involved in fundamental cellular, signaling, and disease processes, and offer valuable opportunities for pharmacological intervention if their properties are fully understood. Taking advantage of the power of yeast genetics and structural biology to direct parallel studies in mammalian cells, we propose: 1. To understand the role of the C-terminal KA-1 domain of MARKlPar-1 kinases in localizing and regulating their in vivo activity. With new structures, kinase assays, and cellular tools in hand, we will also investigate approaches for inhibiting MARK activity by preventing KA-1 domain binding to phospholipids. 2. Having identified core protein-recognition sequences for several yeast PH domains, we will use genetic approaches to test the physiological role of selected PH domain-mediated protein/protein interactions in fundamental cellular processes. In parallel, we will map orthologous interactions in mammalian cells, establishing a new paradigm for PH domains that resembles the role of SH2, and SH3 domains.